/*      $NetBSD: ext2fs.c,v 1.34 2022/04/29 07:42:07 rin Exp $  */

/*
* Copyright (c) 1997 Manuel Bouyer.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
*    notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
*    notice, this list of conditions and the following disclaimer in the
*    documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
* IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/

/*-
* Copyright (c) 1993
*      The Regents of the University of California.  All rights reserved.
*
* This code is derived from software contributed to Berkeley by
* The Mach Operating System project at Carnegie-Mellon University.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
*    notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
*    notice, this list of conditions and the following disclaimer in the
*    documentation and/or other materials provided with the distribution.
* 3. Neither the name of the University nor the names of its contributors
*    may be used to endorse or promote products derived from this software
*    without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
*
* Copyright (c) 1990, 1991 Carnegie Mellon University
* All Rights Reserved.
*
* Author: David Golub
*
* Permission to use, copy, modify and distribute this software and its
* documentation is hereby granted, provided that both the copyright
* notice and this permission notice appear in all copies of the
* software, derivative works or modified versions, and any portions
* thereof, and that both notices appear in supporting documentation.
*
* CARNEGIE MELLON ALLOWS FREE USE OF THIS SOFTWARE IN ITS "AS IS"
* CONDITION.  CARNEGIE MELLON DISCLAIMS ANY LIABILITY OF ANY KIND FOR
* ANY DAMAGES WHATSOEVER RESULTING FROM THE USE OF THIS SOFTWARE.
*
* Carnegie Mellon requests users of this software to return to
*
*  Software Distribution Coordinator  or  [email protected]
*  School of Computer Science
*  Carnegie Mellon University
*  Pittsburgh PA 15213-3890
*
* any improvements or extensions that they make and grant Carnegie the
* rights to redistribute these changes.
*/

/*
*      Stand-alone file reading package for Ext2 file system.
*/

/* #define EXT2FS_DEBUG */

#include <sys/param.h>
#include <sys/time.h>
#include <ufs/ext2fs/ext2fs_dinode.h>
#include <ufs/ext2fs/ext2fs_dir.h>
#include <ufs/ext2fs/ext2fs.h>
#ifdef _STANDALONE
#include <lib/libkern/libkern.h>
#else
#include <string.h>
#endif

#include "stand.h"
#include "ext2fs.h"

#if defined(LIBSA_FS_SINGLECOMPONENT) && !defined(LIBSA_NO_FS_SYMLINK)
#define LIBSA_NO_FS_SYMLINK
#endif

#if defined(LIBSA_NO_TWIDDLE)
#define twiddle()
#endif

#ifndef indp_t
#define indp_t          int32_t
#endif
typedef uint32_t        ino32_t;
#ifndef FSBTODB
#define FSBTODB(fs, indp) EXT2_FSBTODB(fs, indp)
#endif

/*
* To avoid having a lot of filesystem-block sized buffers lurking (which
* could be 32k) we only keep a few entries of the indirect block map.
* With 8k blocks, 2^8 blocks is ~500k so we reread the indirect block
* ~13 times pulling in a 6M kernel.
* The cache size must be smaller than the smallest filesystem block,
* so LN2_IND_CACHE_SZ <= 9 (UFS2 and 4k blocks).
*/
#define LN2_IND_CACHE_SZ        6
#define IND_CACHE_SZ            (1 << LN2_IND_CACHE_SZ)
#define IND_CACHE_MASK          (IND_CACHE_SZ - 1)

/*
* In-core open file.
*/
struct file {
       off_t           f_seekp;        /* seek pointer */
       struct m_ext2fs *f_fs;          /* pointer to super-block */
       struct ext2fs_dinode    f_di;           /* copy of on-disk inode */
       uint            f_nishift;      /* for blocks in indirect block */
       indp_t          f_ind_cache_block;
       indp_t          f_ind_cache[IND_CACHE_SZ];

       char            *f_buf;         /* buffer for data block */
       size_t          f_buf_size;     /* size of data block */
       daddr_t         f_buf_blkno;    /* block number of data block */
};


static int read_inode(ino32_t, struct open_file *);
static int block_map(struct open_file *, indp_t, indp_t *);
static int buf_read_file(struct open_file *, char **, size_t *);
static int search_directory(const char *, int, struct open_file *, ino32_t *);
static int read_sblock(struct open_file *, struct m_ext2fs *);
static int read_gdblock(struct open_file *, struct m_ext2fs *);
#ifdef EXT2FS_DEBUG
static void dump_sblock(struct m_ext2fs *);
#endif

/*
* Read a new inode into a file structure.
*/
static int
read_inode(ino32_t inumber, struct open_file *f)
{
       struct file *fp = (struct file *)f->f_fsdata;
       struct m_ext2fs *fs = fp->f_fs;
       char *buf;
       size_t rsize;
       int rc;
       daddr_t inode_sector;
       struct ext2fs_dinode *dip;

       inode_sector = FSBTODB(fs, ino_to_fsba(fs, inumber));

       /*
        * Read inode and save it.
        */
       buf = fp->f_buf;
       twiddle();
       rc = DEV_STRATEGY(f->f_dev)(f->f_devdata, F_READ,
           inode_sector, fs->e2fs_bsize, buf, &rsize);
       if (rc)
               return rc;
       if (rsize != (size_t)fs->e2fs_bsize)
               return EIO;

       dip = (struct ext2fs_dinode *)(buf +
           EXT2_DINODE_SIZE(fs) * ino_to_fsbo(fs, inumber));
       e2fs_iload(dip, &fp->f_di, EXT2_DINODE_SIZE(fs));

       /*
        * Clear out the old buffers
        */
       fp->f_ind_cache_block = ~0;
       fp->f_buf_blkno = -1;
       return rc;
}

/*
* Given an offset in a file, find the disk block number that
* contains that block.
*/
static int
block_map(struct open_file *f, indp_t file_block, indp_t *disk_block_p)
{
       struct file *fp = (struct file *)f->f_fsdata;
       struct m_ext2fs *fs = fp->f_fs;
       uint level;
       indp_t ind_cache;
       indp_t ind_block_num;
       size_t rsize;
       int rc;
       indp_t *buf = (void *)fp->f_buf;

       /*
        * Index structure of an inode:
        *
        * e2di_blocks[0..EXT2FS_NDADDR-1]
        *              hold block numbers for blocks
        *              0..EXT2FS_NDADDR-1
        *
        * e2di_blocks[EXT2FS_NDADDR+0]
        *              block EXT2FS_NDADDR+0 is the single indirect block
        *              holds block numbers for blocks
        *              EXT2FS_NDADDR .. EXT2FS_NDADDR + EXT2_NINDIR(fs)-1
        *
        * e2di_blocks[EXT2FS_NDADDR+1]
        *              block EXT2FS_NDADDR+1 is the double indirect block
        *              holds block numbers for INDEX blocks for blocks
        *              EXT2FS_NDADDR + EXT2_NINDIR(fs) ..
        *              EXT2FS_NDADDR + EXT2_NINDIR(fs) + EXT2_NINDIR(fs)**2 - 1
        *
        * e2di_blocks[EXT2FS_NDADDR+2]
        *              block EXT2FS_NDADDR+2 is the triple indirect block
        *              holds block numbers for double-indirect
        *              blocks for blocks
        *              EXT2FS_NDADDR + EXT2_NINDIR(fs) + EXT2_NINDIR(fs)**2 ..
        *              EXT2FS_NDADDR + EXT2_NINDIR(fs) + EXT2_NINDIR(fs)**2
        *                      + EXT2_NINDIR(fs)**3 - 1
        */

       if (file_block < EXT2FS_NDADDR) {
               /* Direct block. */
               *disk_block_p = fs2h32(fp->f_di.e2di_blocks[file_block]);
               return 0;
       }

       file_block -= EXT2FS_NDADDR;

       ind_cache = file_block >> LN2_IND_CACHE_SZ;
       if (ind_cache == fp->f_ind_cache_block) {
               *disk_block_p =
                   fs2h32(fp->f_ind_cache[file_block & IND_CACHE_MASK]);
               return 0;
       }

       for (level = 0;;) {
               level += fp->f_nishift;
               if (file_block < (indp_t)1 << level)
                       break;
               if (level > EXT2FS_NIADDR * fp->f_nishift)
                       /* Block number too high */
                       return EFBIG;
               file_block -= (indp_t)1 << level;
       }

       ind_block_num =
           fs2h32(fp->f_di.e2di_blocks[EXT2FS_NDADDR +
           (level / fp->f_nishift - 1)]);

       for (;;) {
               level -= fp->f_nishift;
               if (ind_block_num == 0) {
                       *disk_block_p = 0;      /* missing */
                       return 0;
               }

               twiddle();
               /*
                * If we were feeling brave, we could work out the number
                * of the disk sector and read a single disk sector instead
                * of a filesystem block.
                * However we don't do this very often anyway...
                */
               rc = DEV_STRATEGY(f->f_dev)(f->f_devdata, F_READ,
                       FSBTODB(fp->f_fs, ind_block_num), fs->e2fs_bsize,
                       buf, &rsize);
               if (rc)
                       return rc;
               if (rsize != (size_t)fs->e2fs_bsize)
                       return EIO;
               ind_block_num = fs2h32(buf[file_block >> level]);
               if (level == 0)
                       break;
               file_block &= (1 << level) - 1;
       }

       /* Save the part of the block that contains this sector */
       memcpy(fp->f_ind_cache, &buf[file_block & ~IND_CACHE_MASK],
           IND_CACHE_SZ * sizeof fp->f_ind_cache[0]);
       fp->f_ind_cache_block = ind_cache;

       *disk_block_p = ind_block_num;

       return 0;
}

/*
* Read a portion of a file into an internal buffer.
* Return the location in the buffer and the amount in the buffer.
*/
static int
buf_read_file(struct open_file *f, char **buf_p, size_t *size_p)
{
       struct file *fp = (struct file *)f->f_fsdata;
       struct m_ext2fs *fs = fp->f_fs;
       long off;
       indp_t file_block;
       indp_t disk_block = 0;  /* XXX: gcc */
       size_t block_size, tsz;
       int rc;

       off = ext2_blkoff(fs, fp->f_seekp);
       file_block = ext2_lblkno(fs, fp->f_seekp);
       block_size = fs->e2fs_bsize;    /* no fragment */

       if (file_block != fp->f_buf_blkno) {
               rc = block_map(f, file_block, &disk_block);
               if (rc)
                       return rc;

               if (disk_block == 0) {
                       memset(fp->f_buf, 0, block_size);
                       fp->f_buf_size = block_size;
               } else {
                       twiddle();
                       rc = DEV_STRATEGY(f->f_dev)(f->f_devdata, F_READ,
                               FSBTODB(fs, disk_block),
                               block_size, fp->f_buf, &fp->f_buf_size);
                       if (rc)
                               return rc;
               }

               fp->f_buf_blkno = file_block;
       }

       /*
        * Return address of byte in buffer corresponding to
        * offset, and size of remainder of buffer after that
        * byte.
        */
       *buf_p = fp->f_buf + off;
       *size_p = block_size - off;

       /*
        * But truncate buffer at end of file.
        */
       /* XXX should handle LARGEFILE */
       tsz = (size_t)(fp->f_di.e2di_size - fp->f_seekp);
       if (*size_p > tsz)
               *size_p = tsz;

       return 0;
}

/*
* Search a directory for a name and return its
* inode number.
*/
static int
search_directory(const char *name, int length, struct open_file *f,
       ino32_t *inumber_p)
{
       struct file *fp = (struct file *)f->f_fsdata;
       struct ext2fs_direct *dp;
       struct ext2fs_direct *edp;
       char *buf;
       size_t buf_size;
       int namlen;
       int rc;

       fp->f_seekp = 0;
       /* XXX should handle LARGEFILE */
       while (fp->f_seekp < (off_t)fp->f_di.e2di_size) {
               rc = buf_read_file(f, &buf, &buf_size);
               if (rc)
                       return rc;

               dp = (struct ext2fs_direct *)buf;
               edp = (struct ext2fs_direct *)(buf + buf_size);
               for (; dp < edp;
                   dp = (void *)((char *)dp + fs2h16(dp->e2d_reclen))) {
                       if (fs2h16(dp->e2d_reclen) <= 0)
                               break;
                       if (fs2h32(dp->e2d_ino) == (ino32_t)0)
                               continue;
                       namlen = dp->e2d_namlen;
                       if (namlen == length &&
                           !memcmp(name, dp->e2d_name, length)) {
                               /* found entry */
                               *inumber_p = fs2h32(dp->e2d_ino);
                               return 0;
                       }
               }
               fp->f_seekp += buf_size;
       }
       return ENOENT;
}

int
read_sblock(struct open_file *f, struct m_ext2fs *fs)
{
       static uint8_t sbbuf[SBSIZE];
       struct ext2fs ext2fs;
       size_t buf_size;
       int rc;

       rc = DEV_STRATEGY(f->f_dev)(f->f_devdata, F_READ,
           SBOFF / GETSECSIZE(f), SBSIZE, sbbuf, &buf_size);
       if (rc)
               return rc;

       if (buf_size != SBSIZE)
               return EIO;

       e2fs_sbload((void *)sbbuf, &ext2fs);
       if (ext2fs.e2fs_magic != E2FS_MAGIC)
               return EINVAL;
       if (ext2fs.e2fs_rev > E2FS_REV1 ||
           (ext2fs.e2fs_rev == E2FS_REV1 &&
            (ext2fs.e2fs_first_ino != EXT2_FIRSTINO ||
            (ext2fs.e2fs_inode_size != 128 && ext2fs.e2fs_inode_size != 256) ||
             ext2fs.e2fs_features_incompat & ~EXT2F_INCOMPAT_SUPP))) {
               return ENODEV;
       }

       e2fs_sbload((void *)sbbuf, &fs->e2fs);
       /* compute in-memory m_ext2fs values */
       fs->e2fs_ncg =
           howmany(fs->e2fs.e2fs_bcount - fs->e2fs.e2fs_first_dblock,
           fs->e2fs.e2fs_bpg);
       /* XXX assume hw bsize = 512 */
       fs->e2fs_fsbtodb = fs->e2fs.e2fs_log_bsize + 1;
       fs->e2fs_bsize = MINBSIZE << fs->e2fs.e2fs_log_bsize;
       fs->e2fs_bshift = LOG_MINBSIZE + fs->e2fs.e2fs_log_bsize;
       fs->e2fs_qbmask = fs->e2fs_bsize - 1;
       fs->e2fs_bmask = ~fs->e2fs_qbmask;
       fs->e2fs_ngdb =
           howmany(fs->e2fs_ncg, fs->e2fs_bsize / sizeof(struct ext2_gd));
       fs->e2fs_ipb = fs->e2fs_bsize / ext2fs.e2fs_inode_size;
       fs->e2fs_itpg = fs->e2fs.e2fs_ipg / fs->e2fs_ipb;

       return 0;
}

int
read_gdblock(struct open_file *f, struct m_ext2fs *fs)
{
       struct file *fp = (struct file *)f->f_fsdata;
       size_t rsize;
       uint gdpb;
       int i, rc;

       gdpb = fs->e2fs_bsize / sizeof(struct ext2_gd);

       for (i = 0; i < fs->e2fs_ngdb; i++) {
               rc = DEV_STRATEGY(f->f_dev)(f->f_devdata, F_READ,
                   FSBTODB(fs, fs->e2fs.e2fs_first_dblock +
                   1 /* superblock */ + i),
                   fs->e2fs_bsize, fp->f_buf, &rsize);
               if (rc)
                       return rc;
               if (rsize != (size_t)fs->e2fs_bsize)
                       return EIO;

               e2fs_cgload((struct ext2_gd *)fp->f_buf,
                   &fs->e2fs_gd[i * gdpb],
                   (i == (fs->e2fs_ngdb - 1)) ?
                   (fs->e2fs_ncg - gdpb * i) * sizeof(struct ext2_gd):
                   (size_t)fs->e2fs_bsize);
       }

       return 0;
}


/*
* Open a file.
*/
__compactcall int
ext2fs_open(const char *path, struct open_file *f)
{
#ifndef LIBSA_FS_SINGLECOMPONENT
       const char *cp, *ncp;
       int c;
#endif
       ino32_t inumber;
       struct file *fp;
       struct m_ext2fs *fs;
       int rc;
#ifndef LIBSA_NO_FS_SYMLINK
       ino32_t parent_inumber;
       int nlinks = 0;
       char namebuf[MAXPATHLEN+1];
       char *buf;
#endif

       /* allocate file system specific data structure */
       fp = alloc(sizeof(struct file));
       memset(fp, 0, sizeof(struct file));
       f->f_fsdata = (void *)fp;

       /* allocate space and read super block */
       fs = alloc(sizeof(*fs));
       memset(fs, 0, sizeof(*fs));
       fp->f_fs = fs;
       twiddle();

       rc = read_sblock(f, fs);
       if (rc)
               goto out;

#ifdef EXT2FS_DEBUG
       dump_sblock(fs);
#endif

       /* alloc a block sized buffer used for all fs transfers */
       fp->f_buf = alloc(fs->e2fs_bsize);

       /* read group descriptor blocks */
       fs->e2fs_gd = alloc(sizeof(struct ext2_gd) * fs->e2fs_ncg);
       rc = read_gdblock(f, fs);
       if (rc)
               goto out;

       /*
        * Calculate indirect block levels.
        */
       {
               indp_t mult;
               int ln2;

               /*
                * We note that the number of indirect blocks is always
                * a power of 2.  This lets us use shifts and masks instead
                * of divide and remainder and avoids pulling in the
                * 64bit division routine into the boot code.
                */
               mult = EXT2_NINDIR(fs);
#ifdef DEBUG
               if (!powerof2(mult)) {
                       /* Hummm was't a power of 2 */
                       rc = EINVAL;
                       goto out;
               }
#endif
               for (ln2 = 0; mult != 1; ln2++)
                       mult >>= 1;

               fp->f_nishift = ln2;
       }

       inumber = EXT2_ROOTINO;
       if ((rc = read_inode(inumber, f)) != 0)
               goto out;

#ifndef LIBSA_FS_SINGLECOMPONENT
       cp = path;
       while (*cp) {

               /*
                * Remove extra separators
                */
               while (*cp == '/')
                       cp++;
               if (*cp == '\0')
                       break;

               /*
                * Check that current node is a directory.
                */
               if ((fp->f_di.e2di_mode & EXT2_IFMT) != EXT2_IFDIR) {
                       rc = ENOTDIR;
                       goto out;
               }

               /*
                * Get next component of path name.
                */
               ncp = cp;
               while ((c = *cp) != '\0' && c != '/')
                       cp++;

               /*
                * Look up component in current directory.
                * Save directory inumber in case we find a
                * symbolic link.
                */
#ifndef LIBSA_NO_FS_SYMLINK
               parent_inumber = inumber;
#endif
               rc = search_directory(ncp, cp - ncp, f, &inumber);
               if (rc)
                       goto out;

               /*
                * Open next component.
                */
               if ((rc = read_inode(inumber, f)) != 0)
                       goto out;

#ifndef LIBSA_NO_FS_SYMLINK
               /*
                * Check for symbolic link.
                */
               if ((fp->f_di.e2di_mode & EXT2_IFMT) == EXT2_IFLNK) {
                       /* XXX should handle LARGEFILE */
                       size_t link_len = fp->f_di.e2di_size;
                       size_t len;

                       len = strlen(cp);

                       if (link_len + len > MAXPATHLEN ||
                           ++nlinks > MAXSYMLINKS) {
                               rc = ENOENT;
                               goto out;
                       }

                       memmove(&namebuf[link_len], cp, len + 1);

                       if (link_len < EXT2_MAXSYMLINKLEN) {
                               memcpy(namebuf, fp->f_di.e2di_blocks, link_len);
                       } else {
                               /*
                                * Read file for symbolic link
                                */
                               size_t buf_size;
                               indp_t  disk_block;

                               buf = fp->f_buf;
                               rc = block_map(f, (indp_t)0, &disk_block);
                               if (rc)
                                       goto out;

                               twiddle();
                               rc = DEV_STRATEGY(f->f_dev)(f->f_devdata,
                                       F_READ, FSBTODB(fs, disk_block),
                                       fs->e2fs_bsize, buf, &buf_size);
                               if (rc)
                                       goto out;

                               memcpy(namebuf, buf, link_len);
                       }

                       /*
                        * If relative pathname, restart at parent directory.
                        * If absolute pathname, restart at root.
                        */
                       cp = namebuf;
                       if (*cp != '/')
                               inumber = parent_inumber;
                       else
                               inumber = (ino32_t)EXT2_ROOTINO;

                       if ((rc = read_inode(inumber, f)) != 0)
                               goto out;
               }
#endif  /* !LIBSA_NO_FS_SYMLINK */
       }

       /*
        * Found terminal component.
        */
       rc = 0;

#else /* !LIBSA_FS_SINGLECOMPONENT */

       /* look up component in the current (root) directory */
       rc = search_directory(path, strlen(path), f, &inumber);
       if (rc)
               goto out;

       /* open it */
       rc = read_inode(inumber, f);

#endif /* !LIBSA_FS_SINGLECOMPONENT */

       fp->f_seekp = 0;                /* reset seek pointer */

out:
       if (rc)
               ext2fs_close(f);
       else
               fsmod = "ufs/ext2fs";
       return rc;
}

__compactcall int
ext2fs_close(struct open_file *f)
{
       struct file *fp = (struct file *)f->f_fsdata;

       f->f_fsdata = NULL;
       if (fp == NULL)
               return 0;

       if (fp->f_fs->e2fs_gd)
               dealloc(fp->f_fs->e2fs_gd,
                   sizeof(struct ext2_gd) * fp->f_fs->e2fs_ncg);
       if (fp->f_buf)
               dealloc(fp->f_buf, fp->f_fs->e2fs_bsize);
       dealloc(fp->f_fs, sizeof(*fp->f_fs));
       dealloc(fp, sizeof(struct file));
       return 0;
}

/*
* Copy a portion of a file into kernel memory.
* Cross block boundaries when necessary.
*/
__compactcall int
ext2fs_read(struct open_file *f, void *start, size_t size, size_t *resid)
{
       struct file *fp = (struct file *)f->f_fsdata;
       size_t csize;
       char *buf;
       size_t buf_size;
       int rc = 0;
       char *addr = start;

       while (size != 0) {
               /* XXX should handle LARGEFILE */
               if (fp->f_seekp >= (off_t)fp->f_di.e2di_size)
                       break;

               rc = buf_read_file(f, &buf, &buf_size);
               if (rc)
                       break;

               csize = size;
               if (csize > buf_size)
                       csize = buf_size;

               memcpy(addr, buf, csize);

               fp->f_seekp += csize;
               addr += csize;
               size -= csize;
       }
       if (resid)
               *resid = size;
       return rc;
}

/*
* Not implemented.
*/
#ifndef LIBSA_NO_FS_WRITE
__compactcall int
ext2fs_write(struct open_file *f, void *start, size_t size, size_t *resid)
{

       return EROFS;
}
#endif /* !LIBSA_NO_FS_WRITE */

#ifndef LIBSA_NO_FS_SEEK
__compactcall off_t
ext2fs_seek(struct open_file *f, off_t offset, int where)
{
       struct file *fp = (struct file *)f->f_fsdata;

       switch (where) {
       case SEEK_SET:
               fp->f_seekp = offset;
               break;
       case SEEK_CUR:
               fp->f_seekp += offset;
               break;
       case SEEK_END:
               /* XXX should handle LARGEFILE */
               fp->f_seekp = fp->f_di.e2di_size - offset;
               break;
       default:
               return -1;
       }
       return fp->f_seekp;
}
#endif /* !LIBSA_NO_FS_SEEK */

__compactcall int
ext2fs_stat(struct open_file *f, struct stat *sb)
{
       struct file *fp = (struct file *)f->f_fsdata;

       /* only important stuff */
       memset(sb, 0, sizeof *sb);
       sb->st_mode = fp->f_di.e2di_mode;
       sb->st_uid = fp->f_di.e2di_uid;
       sb->st_gid = fp->f_di.e2di_gid;
       /* XXX should handle LARGEFILE */
       sb->st_size = fp->f_di.e2di_size;
       return 0;
}

#if defined(LIBSA_ENABLE_LS_OP)

#include "ls.h"

static const char    *const typestr[] = {
       "unknown",
       "REG",
       "DIR",
       "CHR",
       "BLK",
       "FIFO",
       "SOCK",
       "LNK"
};

__compactcall void
ext2fs_ls(struct open_file *f, const char *pattern)
{
       struct file *fp = (struct file *)f->f_fsdata;
       size_t block_size = fp->f_fs->e2fs_bsize;
       char *buf;
       size_t buf_size;
       lsentry_t *names = NULL;

       fp->f_seekp = 0;
       while (fp->f_seekp < (off_t)fp->f_di.e2di_size) {
               struct ext2fs_direct  *dp, *edp;
               int rc = buf_read_file(f, &buf, &buf_size);
               if (rc)
                       goto out;
               if (buf_size != block_size || buf_size == 0)
                       goto out;

               dp = (struct ext2fs_direct *)buf;
               edp = (struct ext2fs_direct *)(buf + buf_size);

               for (; dp < edp;
                    dp = (void *)((char *)dp + fs2h16(dp->e2d_reclen))) {
                       const char *t;

                       if (fs2h16(dp->e2d_reclen) <= 0)
                               goto out;

                       if (fs2h32(dp->e2d_ino) == 0)
                               continue;

                       if (dp->e2d_type >= NELEM(typestr) ||
                           !(t = typestr[dp->e2d_type])) {
                               /*
                                * This does not handle "old"
                                * filesystems properly. On little
                                * endian machines, we get a bogus
                                * type name if the namlen matches a
                                * valid type identifier. We could
                                * check if we read namlen "0" and
                                * handle this case specially, if
                                * there were a pressing need...
                                */
                               printf("bad dir entry\n");
                               goto out;
                       }
                       lsadd(&names, pattern, dp->e2d_name,
                           dp->e2d_namlen, fs2h32(dp->e2d_ino), t);
               }
               fp->f_seekp += buf_size;
       }

       lsprint(names);
out:    lsfree(names);
}
#endif

/*
* byte swap functions for big endian machines
* (ext2fs is always little endian)
*
* XXX: We should use src/sys/ufs/ext2fs/ext2fs_bswap.c
*/

/* These functions are only needed if native byte order is not big endian */
#if BYTE_ORDER == BIG_ENDIAN
void
e2fs_sb_bswap(struct ext2fs *old, struct ext2fs *new)
{

       /* preserve unused fields */
       memcpy(new, old, sizeof(struct ext2fs));
       new->e2fs_icount        =       bswap32(old->e2fs_icount);
       new->e2fs_bcount        =       bswap32(old->e2fs_bcount);
       new->e2fs_rbcount       =       bswap32(old->e2fs_rbcount);
       new->e2fs_fbcount       =       bswap32(old->e2fs_fbcount);
       new->e2fs_ficount       =       bswap32(old->e2fs_ficount);
       new->e2fs_first_dblock  =       bswap32(old->e2fs_first_dblock);
       new->e2fs_log_bsize     =       bswap32(old->e2fs_log_bsize);
       new->e2fs_fsize         =       bswap32(old->e2fs_fsize);
       new->e2fs_bpg           =       bswap32(old->e2fs_bpg);
       new->e2fs_fpg           =       bswap32(old->e2fs_fpg);
       new->e2fs_ipg           =       bswap32(old->e2fs_ipg);
       new->e2fs_mtime         =       bswap32(old->e2fs_mtime);
       new->e2fs_wtime         =       bswap32(old->e2fs_wtime);
       new->e2fs_mnt_count     =       bswap16(old->e2fs_mnt_count);
       new->e2fs_max_mnt_count =       bswap16(old->e2fs_max_mnt_count);
       new->e2fs_magic         =       bswap16(old->e2fs_magic);
       new->e2fs_state         =       bswap16(old->e2fs_state);
       new->e2fs_beh           =       bswap16(old->e2fs_beh);
       new->e2fs_minrev        =       bswap16(old->e2fs_minrev);
       new->e2fs_lastfsck      =       bswap32(old->e2fs_lastfsck);
       new->e2fs_fsckintv      =       bswap32(old->e2fs_fsckintv);
       new->e2fs_creator       =       bswap32(old->e2fs_creator);
       new->e2fs_rev           =       bswap32(old->e2fs_rev);
       new->e2fs_ruid          =       bswap16(old->e2fs_ruid);
       new->e2fs_rgid          =       bswap16(old->e2fs_rgid);
       new->e2fs_first_ino     =       bswap32(old->e2fs_first_ino);
       new->e2fs_inode_size    =       bswap16(old->e2fs_inode_size);
       new->e2fs_block_group_nr =      bswap16(old->e2fs_block_group_nr);
       new->e2fs_features_compat =     bswap32(old->e2fs_features_compat);
       new->e2fs_features_incompat =   bswap32(old->e2fs_features_incompat);
       new->e2fs_features_rocompat =   bswap32(old->e2fs_features_rocompat);
       new->e2fs_algo          =       bswap32(old->e2fs_algo);
       new->e2fs_reserved_ngdb =       bswap16(old->e2fs_reserved_ngdb);
}

void e2fs_i_bswap(struct ext2fs_dinode *old, struct ext2fs_dinode *new,
   size_t isize)
{
       /* preserve non-swapped and unused fields */
       memcpy(new, old, isize);

       /* swap what needs to be swapped */
       new->e2di_mode          =       bswap16(old->e2di_mode);
       new->e2di_uid           =       bswap16(old->e2di_uid);
       new->e2di_gid           =       bswap16(old->e2di_gid);
       new->e2di_nlink         =       bswap16(old->e2di_nlink);
       new->e2di_size          =       bswap32(old->e2di_size);
       new->e2di_atime         =       bswap32(old->e2di_atime);
       new->e2di_ctime         =       bswap32(old->e2di_ctime);
       new->e2di_mtime         =       bswap32(old->e2di_mtime);
       new->e2di_dtime         =       bswap32(old->e2di_dtime);
       new->e2di_nblock        =       bswap32(old->e2di_nblock);
       new->e2di_flags         =       bswap32(old->e2di_flags);
       new->e2di_version       =       bswap32(old->e2di_version);
       new->e2di_gen           =       bswap32(old->e2di_gen);
       new->e2di_facl          =       bswap32(old->e2di_facl);
       new->e2di_size_high     =       bswap32(old->e2di_size_high);
       new->e2di_nblock_high   =       bswap16(old->e2di_nblock_high);
       new->e2di_facl_high     =       bswap16(old->e2di_facl_high);
       new->e2di_uid_high      =       bswap16(old->e2di_uid_high);
       new->e2di_gid_high      =       bswap16(old->e2di_gid_high);
       new->e2di_checksum_low  =       bswap16(old->e2di_checksum_low);

       /*
        * Following fields are only supported for inode sizes bigger
        * than the old ext2 one
        */
       if (isize == EXT2_REV0_DINODE_SIZE)
               return;

       new->e2di_extra_isize   = bswap16(old->e2di_extra_isize);
       new->e2di_checksum_high = bswap16(old->e2di_checksum_high);

       /* Following fields are ext4, might not be actually present */
       if (EXT2_DINODE_FITS(new, e2di_ctime_extra, isize))
               new->e2di_ctime_extra   = bswap32(old->e2di_ctime_extra);
       if (EXT2_DINODE_FITS(new, e2di_mtime_extra, isize))
               new->e2di_mtime_extra   = bswap32(old->e2di_mtime_extra);
       if (EXT2_DINODE_FITS(new, e2di_atime_extra, isize))
               new->e2di_atime_extra   = bswap32(old->e2di_atime_extra);
       if (EXT2_DINODE_FITS(new, e2di_crtime, isize))
               new->e2di_crtime        = bswap32(old->e2di_crtime);
       if (EXT2_DINODE_FITS(new, e2di_crtime_extra, isize))
               new->e2di_crtime_extra  = bswap32(old->e2di_crtime_extra);
       if (EXT2_DINODE_FITS(new, e2di_version_high, isize))
               new->e2di_version_high  = bswap32(old->e2di_version_high);
       if (EXT2_DINODE_FITS(new, e2di_projid, isize))
               new->e2di_projid        = bswap32(old->e2di_projid);
}
#endif

#ifdef EXT2FS_DEBUG
void
dump_sblock(struct m_ext2fs *fs)
{

       printf("fs->e2fs.e2fs_bcount = %u\n", fs->e2fs.e2fs_bcount);
       printf("fs->e2fs.e2fs_first_dblock = %u\n", fs->e2fs.e2fs_first_dblock);
       printf("fs->e2fs.e2fs_log_bsize = %u\n", fs->e2fs.e2fs_log_bsize);
       printf("fs->e2fs.e2fs_bpg = %u\n", fs->e2fs.e2fs_bpg);
       printf("fs->e2fs.e2fs_ipg = %u\n", fs->e2fs.e2fs_ipg);
       printf("fs->e2fs.e2fs_magic = 0x%x\n", fs->e2fs.e2fs_magic);
       printf("fs->e2fs.e2fs_rev = %u\n", fs->e2fs.e2fs_rev);

       if (fs->e2fs.e2fs_rev == E2FS_REV1) {
               printf("fs->e2fs.e2fs_first_ino = %u\n",
                   fs->e2fs.e2fs_first_ino);
               printf("fs->e2fs.e2fs_inode_size = %u\n",
                   fs->e2fs.e2fs_inode_size);
               printf("fs->e2fs.e2fs_features_compat = %u\n",
                   fs->e2fs.e2fs_features_compat);
               printf("fs->e2fs.e2fs_features_incompat = %u\n",
                   fs->e2fs.e2fs_features_incompat);
               printf("fs->e2fs.e2fs_features_rocompat = %u\n",
                   fs->e2fs.e2fs_features_rocompat);
               printf("fs->e2fs.e2fs_reserved_ngdb = %u\n",
                   fs->e2fs.e2fs_reserved_ngdb);
       }

       printf("fs->e2fs_bsize = %u\n", fs->e2fs_bsize);
       printf("fs->e2fs_fsbtodb = %u\n", fs->e2fs_fsbtodb);
       printf("fs->e2fs_ncg = %u\n", fs->e2fs_ncg);
       printf("fs->e2fs_ngdb = %u\n", fs->e2fs_ngdb);
       printf("fs->e2fs_ipb = %u\n", fs->e2fs_ipb);
       printf("fs->e2fs_itpg = %u\n", fs->e2fs_itpg);
}
#endif